Journal of Phy8iology (1997), 505.2, pp.529-538

How the thyroid controls metabolism in the rat: different rolesfor triiodothyronine and diiodothyronines

Maria Moreno * t, Antonia Lanni t, Assunta Lombardi t and Fernando Goglia t t

*Dipartimento di Chimica, Universit'a di Salerno, Facolt'a di Scienze, Via S. Allende,84081 Baronissi, Salerno, Italy and t Dipartimento di Fisiologia Generale ed Ambientale,Universita degli Studi di Napoli 'Federico II', Via Mezzocannone 8, I 80134 Napoli, Italy

1. Although the first evidence of a relationship between the thyroid and metabolism wasreported in 1895, the mechanism by which thyroid hormones influence resting metabolicrate in whole animals is still poorly understood. This paper reports an attempt to testwhether diiodothyronines (T2s) and triiodothyronine (T3) have different roles in the controlof resting metabolism (RM).

2. Changes in resting metabolic rate were measured in hypothyroid rats treated acutely (25 ,ug(100 g body weight)-) either with one of the T2s or with T3. Injection of T3 induced anincrease of about 35% in RM that started 25-30 h after the injection and lasted until5-6 days after the injection, the maximal value being observed at 50-75 h. The injection ofTs evoked a temporally different pattern of response. The increases in RM started 6-12 hafter the injection, had almost disappeared after 48 h, and the maximal stimulation wasobserved at 28-30 h.

3. When actinomycin D (an inhibitor of protein synthesis) and T3 were given together, thestimulation of RM was almost completely abolished. The simultaneous injection ofactinomycin D and either of the T s, on the other hand, did not cause any attenuation of thestimulation seen with the T2 s alone.

4. Following chronic treatment (3 weeks) with either T3 or T2s there was a stimulation of organgrowth only after the administration of T3.

5. Chronic administration of either T2s or T3 to hypothyroid rats significantly enhanced theoxidative capacity of each of the tissues considered. In the case of T2s the stimulation wasalmost the same whether it was expressed as an increase in specific activity or total tissueactivity. In the case of T3 the increases were, in the main, secondary to the hypertrophic orhyperplastic effect.

6. These results indicate that T2s and T3 exert different effects on RM. The effects of T2s arerapid and possibly mediated by their direct interaction with mitochondria. Those of T3 areslower and more prolonged, and at least partly attributable to a modulation of the cellularityof tissues that are metabolically very active.

Among the most pronounced physiological effects attributedto thyroid hormones in many adult endothermic vertebratesis their control over metabolism. Although the first evidenceof an increase in metabolic rate in subjects given thyroidextract was reported as long ago as 1895 by Magnus-Levy,the literature contains very few papers dealing with theaction of thyroxine (T4) and T3 on metabolic rate of wholeanimals. Furthermore, although more than 30 years haveelapsed since the appearance of the articles by Tata (Tata,Ernster & Lindberg, 1962; Tata, 1963) on the control ofbasal metabolic rate by thyroid hormones, the mechanismby which thyroid hormones influence basal or resting

metabolic rate in whole animals (also called the calorigeniceffect) is very poorly understood.

Several hypotheses have been proposed to explain thecalorigenic effect of thyroid hormones, but none has receiveduniversal acceptance (Sestoft, 1980). Indeed, conflictingresults are to be found in the literature and a debate is stillin progress as to whether the cellular target for the earlyaction of thyroid hormones on energy metabolism is thenucleus or the mitochondria (for review, see Soboll, 1993). Tosome extent, the controversies may be a consequence of theuse of a wide variety of animal models in the various

I To whom correspondence should be addressed.

7060 529

) by guest on February 25, 2013jp.physoc.orgDownloaded from J Physiol (

M. Moreno, A. Lanni, A. Lombardi and F Goglia

investigations of the actions of thyroid hormones. Indeed,the use of acute, as opposed to chronic, hormonal treatmentand the use of different methods to induce hypothyroidismcould have led different authors towards differentinterpretations (Lanni, Moreno, Lombardi & Goglia, 1996).In fact, studies on the characteristics of deiodinase enzymes(Kohrle, 1994) have revealed that very different animalmodels are produced depending on whether the hypo-thyroidism is induced by surgical or chemical thyroidectomy(Lanni et al. 1996).

In the past, the iodothyronines other than T4 and T3 thatare present in biological fluids have been regarded asinactive. Consequently, studies of the calorigenic effect ofthyroid hormones have focused almost exclusively on T4 andT3. Indeed, of the thyroid hormones, T3 has beenuniversally considered to be the active form. Recently,however, a growing number of studies have indicated thattwo diiodothyronines, 3,3'-diiodo-L-thyronine (3,3'-T2) and3,5-diiodo-L-thyronine (3,5-T2) (together referred to as T2s)could be of biological relevance. These iodothyronines areable rapidly to stimulate the oxidative capacity andrespiration rate of rat liver mitochondria (Lanni, Moreno,Cioffi & Goglia, 1992, 1993; O'Reilly & Murphy, 1992;Lanni, Moreno, Lombardi & Goglia, 1994b) and these effectscould be mediated by the binding of T2s to specificmitochondrial sites (Goglia, Lanni, Horst, Moreno & Thoma,1994; Lanni, Moreno, Horst, Lombardi & Goglia, 1994a).Recently, we reported that T2s induce a dose-dependentcalorigenic effect when chronically injected intohypothyroid rats (Lanni et al. 1996). In such animals,following the administration of T2s, we found a goodcorrespondence between the increase in resting metabolismand the increase in the specific oxidative capacity (expressedas cytochrome oxidase (COX) activity in (ng atoms 0) min'(mg protein)-') of tissues that are metabolically very active,such as liver, gastrocnemius muscle, heart and brownadipose tissue (BAT). However, a similar closecorrespondence could not be shown in hypothyroid ratsinjected with T3. In fact, while the administration of T3increased their resting metabolism to euthyroid values orabove, a relevant stimulatory effect on specific oxidativecapacity at the tissue level was observed only in the liver.This could not account for the observed increase in RM.This apparent discrepancy could be explained if we assumethat the effects of T2s and T3 on RM are mediated bydifferent mechanisms at the cellular level. Thus, the effect ofT3 could be mediated via a nuclear pathway, resulting in ahypertrophic or hyperplastic effect on tissues that aremetabolically very active. In contrast, if T2 s havemitochondria as their target, this could result in a directand more rapid stimulation of RM, without an effect onorgan mass.

We have tested these hypotheses by: (a) treating ratsacutely either with T2s or with T3 and measuring the timeof onset and following the time course of the calorigeniceffect (measured as the effect on resting metabolism in the

whole animal); and (b) treating rats chronically with thesame substances and following the development of thehypertrophic or hyperplastic effect (in the liver, gastro-cnemius muscle, brown adipose tissue and heart) bydetermining the mass of the tissues as well as their totalprotein and DNA content.

As a model for both acute and chronic studies, we usedhypothyroid animals in which hypothyroidism was inducedby combined treatment with propylthiouracil (PTU) andiopanoic acid (IOP). This treatment, which induces a severehypothyroidism and, at the same time, inhibits all the threeknown types of deiodinase enzyme, permits us to attributethe observed effects to the iodothyronines injected, ratherthan to any of their deiodinated products. Moreover, inhypothyroid rats the presence of endogenous iodothyroninesis strongly reduced and the endogenous occupation of thebinding sites should consequently be lowered.

A preliminary version of these results has been published inabstract form (Moreno, Lanni, Lombardi & Goglia, 1996).

METHODSAnimals and treatmentsMale Wistar rats (250-300 g) living in a temperature-controlledroom at 28 °C were kept one per cage under an artificial lightingregime of 12 h light: 12 h darkness. A commercial mash (Charles-Rivers, Lecco, Italy) was available ad libitum and the animals hadfree access to water. All experiments were performed in accordancewith local and national guidelines covering animal experiments. Atthe end of the experiment, the rats were anaesthetized by i.P.administration of chloral hydrate (40 mg (100 g body weight(BW))-') and killed by decapitation. Hypothyroidism was inducedin some rats by the administration of PTU (0-1% w/v in drinkingwater for 3 weeks) together with IOP (6 mg (100 g BW)-f) (Lanni etal. 1996). These rats are referred as the 'P + I' group.

To enable determination of the time course of the calorigenic effectof iodothyronines, hypothyroid animals were acutely injected i.P.with either (i) a single dose of 25 jug (100 g BW)-1 of either T3 or3,5-T2, or (ii) in the case of 3,3'-T2, three doses of25 jug (100 g BW)-' at 12 h intervals. As a control for the possibleeffects of the injection itself, saline was injected i.P. intohypothyroid control rats. The dose of 25 ug (100 g BW)f' was usedbecause it produces a clear-cut effect on RM and indeed restores itto the level observed in euthyroid control rats (Tata, Ernster,Lindberg, Arrhenius, Pedersen & Hedman, 1963). This dose givenacutely is not a large dose; in fact, it has been established that200 #g (100 g BW)-1, given i.v., is the acute dose needed to obtain95% nuclear receptor saturation for 24 h (Jump, Narayan, Towle

& Oppenheimer, 1984).

It is conceivable that the administration of PTU and IOP and theconsequent alteration in iodothyronine metabolism could inducenon-physiological rather than physiological metabolic effects.Actually, it is known that the substances we used to inducehypothyroidism have no effect on metabolism in man (Acheson &Burger, 1980). Nevertheless, to be sure that these two substances donot induce any non-physiological effects in rats, we also measuredthe effect of the acute administration of 25 #sg iodothyronines(100 g BW)-' on RM in two other control groups of rats in whichhypothyroidism was induced either by surgical thyroidectomy

530 J Physiol.505.2

) by guest on February 25, 2013jp.physoc.orgDownloaded from J Physiol (

Iodothyronines and metabolic rate

without administration of P + I (referred to as Tx) or bythyroidectomy and concomitant P + I administration (referred toas Tx + P + I). A similar effect on RM in all the hypothyroidgroups, regardless of the way hypothyroidism was induced, wouldindicate that non-physiological effects were not induced in animalstreated with P + I.

To enable evaluation of the possible involvement of changes inprotein synthesis in the action of the iodothyronines tested,actinomycin D (8 jug (100 g BW)-') was injected in combinationwith 25 jg (100 g BW)-1 of 3,5-T2 or T3. At the extremely lowdoses used here, actinomycin D only inhibits the synthesis ofmessenger RNA (Goldberg & Rabinowitz, 1962).

To enable the hypertrophic or hyperplastic effect of iodothyronineson the tissues to be followed, P + I rats received, by a once-dailyi.P. injection over a 3 week period, one of three different doses (2f5,5 or 10 jug (100 g BW)-') of either T3, 3,5-T2 or 3,3'-T2. Thismethod also allowed us to plot a dose-response relationship. Theabove doses fall within the range of doses used in a previous studyof dose dependency (Oppenheimer, Schwartz, Lane & Thompson,1991). At the end of the treatment, each rat was killed, the trunkblood collected and the serum separated and stored at -20°C forlater measurement of the concentration of T3 and T4. InterscapularBAT, heart, gastrocnemius muscle and liver were dissected out,cleaned, immediately weighed (wet weight) and processed for thedetermination of COX activity, protein content and DNA content.As an index of cellularity, we adopted the protein/DNA ratiowhich, following an increase in organ weight, indicates a hyper-trophy if it increases, but a hyperplasy if it decreases or remainsunchanged.

Resting metabolismThe resting metabolism was measured using open-circuit indirectcalorimetry. The rat was placed in a respiration chamber(-32 cm x 20 cm x 19 cm) with airflow being measured using anO2-ECO mass flow controller (Columbus Instruments InternationalCorporation, Columbus, OH, USA). Details of this set-up and of theway that measurements are made have been given by Lanni et al.(1996). The measurements for the calculation of RM (the lowestmetabolic rate of a resting animal when it is not in a post-absorptive or fasting state and is not sleeping) were taken at 28 °Cbetween 11.00 h and 16.00 h, when the energy expenditure was ata lower level than in any other period of the day. Measurementswere taken before and at various intervals after the rat was injectedwith T3 or with one of the T2s (see above).

Analytical proceduresCytochrome oxidase (COX) activity was determined polaro-graphically at 25 °C, using a Clark oxygen electrode and amodification (Barre, Bailly & Rouanet, 1987) of the procedure ofAulie & Grav (1979). This required 1'5 ml of reaction mediumcontaining 30 jM cytochrome c, 4#M rotenone, 0 5 mm dinitro-phenol (DNP), 10 mm sodium malonate and 75 mm Hepes buffer, atpH 7f4. Samples of liver, skeletal muscle, brown adipose tissue andheart were finely minced, diluted 1/10 (w/v) and homogenized inmodified Chappel-Perry medium (mM): 1 ATP, 50 Hepes bufferadjusted to pH 7*4, 100 KCl, 5 MgCl2, 1 EDTA and 5 EGTA. Thehomogenate was then diluted 1:2 (v/v) in the same medium, withlubrol (100 mg (g tissue)-') being added to unmask the enzymeactivity of the tissue. It was then left standing in ice for 30 min.

Cytochrome oxidase activity was measured as the differencebetween (i) the rate of oxygen consumption observed after theaddition of substrate (4 mm sodium ascorbate with 0 3 mM NN\-N',N'-tetramethyl-p-phenylene-diamine (TMPD)) and homogenate

and (ii) the rate of oxygen consumption observed after the additionof substrate alone. This method took into account the auto-oxidation of ascorbate.

For the measurement of the activities of the deiodinases, tissuesamples were individually homogenized on ice in 0-25 M sucroseand 10 mm Hepes (pH 7 0) containing 10 mm DTT ( dithiothreitol).Liver and brain microsomes for type I deiodinase (ID-I) activityand type III deiodinase (ID-III) activity, respectively, wereobtained as described by Visser, Kaptein, Terpstra & Krenning,(1988). BAT infranatants, for type II deiodinase (ID-II) activity,were obtained as described by Leonard, Mellen & Larsen (1983).They were all immediately frozen in a dry-ice-acetone bath andstored at -80 °C until assay.

Samples of reverse [3',5'-125I]T3 ([3',5'-125I]rT3) and [5'-1251]T3 wereprepared by radioiodination of 3,3'-T2 or 3,5-T2 (Henning, Berlin,Germany), respectively, using the chloramine-T method asdescribed by others (Visser, Docter & Hennemann, 1977; Visser,Krieger-Quist, Docter & Hennemann, 1978). Labelled productswere purified by Sephadex LH-20 chromatography (PharmaciaUppsala, Sweden) and purity was checked by HPLC analysis usingunlabelled compounds as references. Free 125I was eliminated from[3',5'- 251]rT3 and [5'-'251]T3 on Sephadex LH-20 immediatelybefore each experiment. Iodothyronines were obtained fromMMDRI, Henning Berlin R&D (Berlin, Germany). [1251]-Na forradioiodination was purchased from Amersham.

The essential step in the assay of type I deiodinase activity was theproduction of radioiodide from [3',5'-125I]rT3 by ID-I deiodinase. A2 jug sample of liver microsomal protein was incubated for 30 minat 37°C with 01 jum rT3 and O-100 000 c.p.m. [3',5'-_25I]rT3 in200 jul of 0'2 M phosphate buffer (pH 7 2) with 4 mm EDTA and5 mM DTT. Incubations were performed in triplicate in a shakingwaterbath and the reactions were stopped by the addition of 100 ul5% bovine serum albumin at 0 'C. Protein-bound iodothyronineswere precipitated by the addition of 500 jul of 10% trichloroaceticacid. After incubation of the mixtures at 0 'C, they werecentrifuged and the radioactivity in the supernatant was thendetermined. Enzymatic deiodination was corrected for non-enzymatic 1251 production (as determined in blank incubationswithout microsomes) and multiplied by 2 to account for randomlabelling and the deiodination of the 3' and 5' positions in [3',5'-125I]rT3.The activity of the type II deiodinase enzyme was measured inBAT according to the method of Leonard et al. (1983); this involvesmeasuring the release of radioiodide from [3',5'-125I]rT3. A 20 jugsample of BAT infranatant proteins (obtained by centrifugation ofBAT homogenate at 500 x g for 10 min) was incubated in triplicatefor 60 min at 37 'C in a shaking waterbath with 2 nM rT3 and--100 000 c.p.m. of [3',5'-125I]rT3 in 200 jul 0 1 M phosphate buffer(pH 7 2) with 2 mm EDTA and 20 mm DTT. The reactions werestopped by the addition of 100 jul 5% BSA and the 1251 producedwas isolated and analysed as described above. Random labellingand deiodination of the 3' and 5' positions of [3',5'-525I]rT3 wastaken into account in the calculation of ID-II activity.

Type III deiodinase activity in brain microsomes was determinedby measuring the formation of 3[3'-125I]T2 from [5'-_251]T3 byHPLC analysis as reported by Schoenmakers, Pigmans & Visser(1995). A 100 jug sample of brain microsomal protein was incubatedin triplicate for 60 min at 37 'C in a shaking waterbath with 1 nMT3 and _100000 c.p.m. of [5'-125I]T3 in 200 ju1 of 0 1 M phosphatebuffer (pH 7'2) with 4 mm EDTA and 10 mm DTT. The reactionswere stopped by the addition of 300 jul methanol on ice. Aftercentrifugation of precipitated proteins, the supernatants were

J Physiol.505.2 531

) by guest on February 25, 2013jp.physoc.orgDownloaded from J Physiol (

M. Moreno, A. Lanni, A. Lombardi and F Goglia

Table 1. Total and free T3 and total T4 serum levels in euthyroid (N), hypothyroid (P + I) andhypothyroid rats chronically treated with T3 (P + I + T3)

Total Free TotalGroup serum T3 serum T3 serum T4

(nmol l-l) (pmol lFl) (nmol F-l)

N 0-88 + 006b 4-1 + 0,4b 62-0 + 3 0bP + I 0'17 + 0'02a 0.5 + 0.05a 7-3 + 0,8aP + I + T3 (2 5 jug T3 (100 g BW)-') 1P11 + 0.11 b 3-6 + 0,3b 6'8 + 0.6aP + I + T3 (5 jg T3 (100 g BW)-') 1P28 + 0 14b 4-9 + 0,5bc 6-5 + 0.5aP + I + T3 (10jug T3 (100 g BWY')l 176 + 0.16c 5.8 + 0.7c 6-1 + 0.5a

Results are the means + S.E.M. of 5 experiments for each group, each experiment being performed intriplicate. Values labelled with different letters are significantly different (P < 0 05) from each other.

analysed for 3[3'-_25I]T2 formation by HPLC analysis involvingelution with a 45:50 (v/v) mixture of methanol and 20 mMammonium acetate (pH 4 0) at a flow of 0-8 ml min-'.Serum total T4 and T3 levels and free T3 were determined insamples of serum using reagents and protocol supplied by Becton-Dickinson (Orangeburg, NJ, USA).

The protein concentration was determined by the method ofHartree (1972) using bovine serum albumin as standard. The DNAcontent was measured by a colorimetric method (Burton, 1956).

Results are expressed as means + S.E.M. The statistical significanceof differences between groups was determined by a one-wayanalysis of variance followed by a Student-Newman-Keuls test.Comparison between independent means was performed using aStudent's t test.

RESULTSTotal and free T3 and total T4 concentrationsThe combined administration of PTU and IOP producesrats with severe hypothyroidism. Indeed, total and free T3and total T4 levels were significantly lower in suchhypothyroid rats than in euthyroid ones (by about 80, 92and 88%, respectively; Table 1). In Tx and in Tx + P + Irats, the reduction in all three levels was of the same orderas that seen in P + I animals (data not shown).

Following the chronic injection of various doses of T3 toP + I rats, the total and free T3 levels were (i) within therange of values observed in euthyroid animals (at a dose of2-5 ,sg (100 g BW)-'), (ii) slightly higher (at a dose of 5 ,ug(100 g BW)-') or (iii) significantly higher (at a dose of 10 ,ug(100 g BW)-') (see Table 1). Our values for these circulatinglevels in euthyroid, hypothyroid and hypothyroid + T3-treated animals are in agreement with values previouslyreported in rats under similar conditions (Francavilla et al.1991).

Deiodinase activity in hypothyroid ratsThe activities of the deiodinase enzymes in P + I animalswere either completely blocked or, in the case of the type IIdeiodinase, greatly reduced (by 66 %) (Table 2).

Similar results were obtained in Tx + P + I animals, but inTX animals the activities were differently affected. In thelatter animals, type I and type III deiodinases wereinhibited by about 63% and 67 %, respectively, while typeII activity was increased nearly 4-fold (Table 2).

Effect of acute injection of iodothyronines on the RMof hypothyroid ratsRM was considerably lower in all the hypothyroid groupsthan in euthyroid controls (0 94 + 0-02, 0 90 + 0'02,0'89 + 0-03 and lP45 + 003102 (kg075)- h-,in P + I, Tx,Tx + P + I and euthyroid control rats, respectively), thusconfirming the hypothyroid status of the animals. Figure 1illustrates the effect of the acute administration of either3,5-T2 or T3 on the RM of P + I rats. The injection of T3caused an increase of about 35% in resting metabolic rate;the increase started 25-30 h after the injection, reachedmaximal values at 50-75 h and lasted until 5-6 days afterthe injection (Fig. 1, 0). This trend was in accordance withthat previously observed by Tata in his early studies onthyroidectomized rats (Tata et al. 1962; Tata, 1963).

The injection of T2s, on the other hand, evoked a temporallydifferent pattern of response. The injection of 3,5-T2 causedan increase of about 40% in RM (Fig. 1, *). However, inthis case the increase started between 6 and 12 h after theinjection, peaked at about 30 h and had almost disappearedat 48 h. A similar trend was observed in the response to3,3'-T2 (data not shown), but in that case we needed to giveat least three injections 12 h apart to produce the effect.With 3,3'-T2, the increase in RM started 6-12 h after thelast injection and the response had almost disappeared by30 h after the last injection (not shown). As the RM actuallydecreased during the first few hours of treatment in the caseof animals injected with 3,3'-T2, the increase was by about17% if referred to the initial RM value (before thebeginning of the treatment), but by about 27% if referredto the RM value measured immediately before the lastinjection.The simultaneous injection of actinomycin D and either ofthe T2s did not cause any attenuation of the stimulation

532 J Physiol.505.2

) by guest on February 25, 2013jp.physoc.orgDownloaded from J Physiol (

Iodothyronines and metabolic rate

Table 2. Activity of liver ID-I, BAT ID-II and brain ID-III in hypothyroid (P + I, Tx + P + Iand Tx) and euthyroid (N) rats

Group ID-I activity ID-II activity ID-III activity(pmol I (fmol I (fmol 3,3'-T2

(min (mg protein))-') (h (mg protein))-) (min (mg protein))-)

P + I Undetectable 14 + 1 * UndetectableTx + P + I Undetectable 13 + 1 UndetectableTx 75+5* 153+14* 9+1*N 202 + 12 41 +3 27 + 1

Results are the means + S.E.M. of 5 experiments for each group, each experiment being performed intriplicate. * Values significantly different (P< 0 05) from N rats.

seen with the T2s alone (Fig. 1, O). In contrast, whenactinomycin D and T3 were given together, the stimulationofRM by T3 was almost completely abolished (Fig. 1, 0).

The effects of the T2s and T3 on RM were almost the samein Tx animals as in Tx + P + I animals and none of theseeffects was substantially different from the correspondingeffect in the P + I group. However, one difference wasobserved between Tx animals and the others and layessentially in the steepness of the curves showing changes inRM as a function of time. Anyhow, in Tx rats: (i) the peakincrease in RM induced by T2s preceded that induced by T3by 26-36 h and (ii) the maximal percentage changesfollowing iodothyronines administration were of the sameorder (+27-35% in the case of T2 and +36% in the case ofT3).

It should be emphasized that the deiodinase activities weremarkedly altered in Tx animals as well as in P + I andTx + P + I animals, but in a different way. In fact, in Txrats ID-I and ID-III were inhibited, while I-DII wasactivated (Table 2). Further, in P + I and in Tx + P + I ratsthese activities (due to the inhibition exerted by PTU andIOP) remain unchanged by administration of iodothyronines.

Figure 1. Changes in the resting metabolic rateof hypothyroid rats following administrationof iodothyronines with and withoutactimomycin DHypothyroidism was induced by combinedtreatment with PTU and IOP (P + I). The dose was25 isg (100 g BW)-1 for both triiodothyronine (T3)and 3,5-diiodo-L-thyronine (3,5-T2).*, 3,5-T2alone; O, 3,5-T2 +actinomycin D; *, T3alone;0, T3 + actinomycin D. Each data point shows themean + S.E.M. from 5 rats. The values are expressedas the percentage change from the value at time 0(i.e. immediately before the injection).

._c2e

0Ea

Co4)0CoCu0

co>

0C0

._

a.0)

By contrast, it is known that in Tx rats the deiodinaseactivities are further altered following T3 administration,and in a way that tends to reverse the effect ofthyroidectomy (ID-I and ID-III are stimulated (Kaplan &Utiger, 1978; Esfandiari, Courtin, Lennon, Gavaret &Pierre, 1992) while ID-II is inhibited (Leonard, Kaplan,Visser, Silva & Larsen, 1981)). Consequently, using a Txmodel we have a pattern of deiodination that is differentbefore and after T3 administration (this alteration mayunderlie the aforementioned slight differences in thesteepness of the curves observed in Tx animals). This beingso, and as stated in the Introduction, it becomes problematicto attribute an observed effect to a given iodothyroninerather than to its possible metabolic product (i.e. T3 ratherthan T2). Having considered all the above, we decided thatour chronic studies with different doses of iodothyroninesshould be performed only on P + I rats. Our main reasonswere: (a) the acute effect of iodothyronines were almost thesame in the various hypothyroid conditions, (b) in P + I ratsthe deiodinase activities are always at the same level and(c) we could thus avoid imposing another form of stress(surgical thyroidectomy).

Time (h)

J Physiot 505.2 533

) by guest on February 25, 2013jp.physoc.orgDownloaded from J Physiol (

M. Moreno, A. Lanni, A. Lombardi and F Goglia

Table 3. Specific (SCOX) and total (TCOX) cytochrome oxidase activity from hypothyroid rats andfrom hypothyroid rats treated with various iodothyronines

Liver Gastrocnemius Heart BAT

Group SCOX TCOX SCOX TCOX SCOX TCOX SCOX TCOX

P+I 353+30 404+32 230+13 35+2-8 1734+168 84+6-7 1442+109 10.0+0 9

P+I+3,5-T22-5 /sg 402+ 39(14) 504±40(25) 302 ± 22*(32) 42 +4-2(20) 1790±171(3) 83 ±8-2(0) 1779±121(23) 10-7±0-9(7)5 /sg 477 ± 6*(24) 696 ± 60*(72) 304+ 21*(32) 54 ± 43*(54) 2419 + 212(40) 119±12-2(42) 1989±159*(38) 14-8+1-5*(48)10 #sg 570_53*(61) 651 ± 40*(61) 334+ 33*(45) 54 + 50*(54) 2425+ 221(40) 135±15-3*(61) 2307±187*(60) 20-2+2.1*(100)

P+I+3,3'-T22'5jug 416±33(17) 488±20(21) 303+30*(32) 46±4-2(31) 1954±160(13) 88±7'9(5) 1476±122(2) 11P5±1P1(15)5 jug 423±19(20) 537 ± 22*(32) 359+ 10*(56) 63 +50*(80) 2433 +197(40) 126 ±13-0*(50) 1869+169(30) 17-3+1.4*(70)10 #sg 469+ 13*(32) 510+ 20*(23) 376+ 35*(63) 51 +4.8*(46) 2421 + 213(40) 116+12-1(38) 2173+132*(51) 14-7+112*(47)

P+I+T32-55,g 473 +45(34) 789+ 80*(95) 265+30(15) 52 + 4.8*(49) 1800+177(4) 153+13.2*(82) 1668+103(16) 22-8+223*(128)5 /tg 508 + 48*(44) 749 + 75*(85) 280 ± 25(22) 62 + 5 0*(77) 1850 + 200(7) 134+ 12-7*(60) 1586 + 152(10) 18-4+ 1.5*(84)10jug 614+58*(74) 1060 +100*(162) 290 +10(26) 67 + 7-0*(91) 2048+183(18) 165+ 14 2*(96) 1658+127(15) 17-3+1 4*(73)

SCOX is expressed as (ng atoms 0) min-' (mg protein)-'; TCOX is expressed as (jsg atoms 0) min-. Resultsare presented as means + S.E.M. Values in parentheses represent the percentage increase over P + I value.* Values significantly different (P< 0 05) from P + I rats. n = 9 for P + I group, n = 5 for eachhypothyroid iodothyronine-treated group, where n is no of experiments. Groups given different doses of3,5-T2, 3,3'-T2 and T3 are distinguished by the dose which is given per 100 g BW under the group names inthe left column.

Effect of chronic treatment with iodothyronines onspecific and total tissue COX activity of liver, gastro-cnemius muscle, BAT and heartThe values for the specific (SCOX) and total oxidative(TCOX) capacity of the various tissues are reported inTable 3. An analysis of variance revealed that, in the case of3,3'-T2 and 3,5-T2, hormonal treatment was effective instimulating both specific and total COX activity (P< 0 05;F > 3 72). The effect of the administration of T2s was dosedependent, the maximal effect occurring at a dose of10 4tg (100 g BW)-'. Administration of 3,5-T2 to hypo-thyroid rats at a dose of 10 Ag (100 g BW)-' stimulated thespecific oxidative capacity of the tissues by 40-61%, theeffects on heart and muscle (40 and 45 %, respectively) beingweaker than those on BAT and liver (60 and 61 %,respectively). Administration of 3,3'-T2 to hypothyroid ratsat a dose of 10 jug (100 g BW)-' stimulated the specificoxidative capacity of the tissues by 32-63 %, the effects onliver and heart (32 and 40%, respectively) being weakerthan those on BAT and muscle (51 and 63%, respectively).When the oxidative capacity was expressed as total activity,a similar pattern was observed. In percentage terms, in fact,the stimulation of COX activity by T2s was almost the samewhether the activity was expressed as specific or totalactivity.

In the case of T3, an analysis of variance revealed thathormonal treatment was effective in stimulating specific

activity only in the liver (P < 0-02; F= 7'12). Actually, thelack of an effect of T3 on specific COX activity in BAT is notsurprising since it has been shown that BAT preferentiallyutilizes plasma T4 and that it only poorly utilizes T3 derivedfrom the plasma and extracellular space (Bianco & Silva,1987). However, when the oxidative capacity was expressedas total activity, the stimulation was evident and significantin all organs and tissues tested. In percentage terms(Table 3; values in parentheses), the changes from P + Ivalues were much greater when expressed in terms of totalactivity than when expressed in terms of specific activity.At a dose of 10 #sg (100 g BW)-1, in fact, expressing theincreases in COX activity in terms of total activity (ratherthan in terms of specific activity) caused the size of theincrease to be more than doubled in the liver (162% vs.74%), 3-5 times greater in muscle (91 % vs. 26 %), 5-5 timesgreater in the heart (96% vs. 18%) and about 5 timesgreater in BAT (73% vs. 15%).

Effect of chronic treatment with iodothyronines onbody weight, organ mass, tissue protein content andtissue DNA content in hypothyroid ratsBody mass and the mass of individual organs are significantlylower in hypothyroid rats than in euthyroid ones (Table 4).Following the chronic administration of iodothyronines, allthe organs examined were significantly heavier in T3-treatedanimals than in hypothyroid ones (see Table 4), the greatesteffects being observed in BAT and heart. However, no

J Physiol.505.2534

) by guest on February 25, 2013jp.physoc.orgDownloaded from J Physiol (

Iodothyronines and metabolic rate

Table 4. Body and organ masses from euthyroid (N), hypothyroid (P + I) and triiodothyronine-treated hypothyroid (P + I + T3) rats

Group n Body mass Liver Gastrocnemius BAT Heart(g) (g) (g) (mg) (mg)

P+I 9 263+8t 9-2+0-6t 1P56+0-08t 176_lt 644±33t(35-1 + 1P9) (5'94 + 0'29) (0'68 + 0 05) (2-44 ± 0 09)

N 7 312+10 12-5+07 1-86+0-08 263+14 870+32(40 0 + 2 0) (5 9 + 0 04) (0-84 + 0 05) (2-80 + 0 50)

P + I + T32-5 ug 5 305+15* 11P4+01 1P82+0-12 281+46* 1125+15*

(37-7 + 1-4) (5-96 + 0-01) (0-91 + 004)* (3-74+ 051)*5 jg 5 282 + 15 11P1 + 0-6 1.95 + 0.02* 359 + 24* 973 + 28*

(39-5 + 0-5) (6-93 + 033) (1P29 + 0o16)* (3-45 + 021)*10jug 5 306 + 6* 12.8 + 1.8* 2.13 + 0.13* 446 + 63* 1137 + 33*

(41P7 ± 2 0) (6-94 + 0 53) (1-41 + 0.21)* (3 7 ± 0 08)*

Values in parentheses represent tissue/body weight ratio x 103. Results are presented as means + S.E.M.;n, no. of experiments. * Values significantly different (P< 0 05) from P + I rats; t values significantlydifferent (P< 0'05) from N rats. Because neither 3,3'-T2 nor 3,5-T2 treatment had a significant effect onthe above parameters (treated vs. hypothyroid rats), the data are not reported. Different P + I + T3 groupsare distinguished by the doses of T3 which are given per 100 g BW under the group name in the leftcolumn.

Table 5. Total protein content (mg), total DNA content (mg) and protein/DNA ratio in liver,gastroonemius muscle, BAT and heart homogenates from euthyroid (N), hypothyroid (P + I) andtriiodothyronine-treated hypothyroid (P + I + T3) rats

P+I P+I+T3 N

2-5 juFg 5 sg 10 jg(n=9) (n=5) (n=5) (n=5) (n=7)

Liver Protein 1138 + 73-1t 1457 + 70* 1553 + 68* 1745 + 153* 1529 + 116DNA 36+3-5 38+0-5 36+2-8 35+2-9 36+19Protein/DNA 31 + 2-7t 38 + 1P3 43 + 3.4* 50 + 4.5* 42 + 4 00

Gastrocnemius Protein 170 + 15 192 + 2 196 + 16 204 + 19 195 + 2 *DNA 3-6+003 4.7+0.1* 5.2+0.4* 6.2+0.6* 4-8+003Protein/DNA 47 + 4-0 41 +1P3 38 + 2-7 33 + 2.7* 41 + 3 9

BAT Protein 6-7+006 13.7+0.7* 11.6+1.2* 10.3+2.2* 9.4+0.1*DNA 0-38 + 0-03 0-84 + 0-01* 098 + 0.03* 1-28 + 0.16* 0-51 + 0-05Protein/DNA 18+ 10 16+0-8 12+0.9* 8+ 1.5* 18+ 1-6

Heart Protein 54 + 3-8 65 + 7-5 78 + 6.8* 106 + 10.1* 56-4 + 4-3DNA 2-1 +0i 2 2-5 + 0-2 2-7 + 0-3 3-2 + 0-4 2-6 + 0-2Protein/DNA 25 + 2-2 27 + 3-4 29 + 2-7 33 + 3,4 22 + 2-0

Because neither 3,3'-T2 nor 3,5-T2 treatment had a significant effect on the above parameters (treated vs.hypothyroid rats), the data are not reported. Results are presented as means + S.E.M.; n, no. ofexperiments. * Values significantly different (P< 0-05) from P + I rats; t values significantly different(P< 0 05) from N rats. Different P + I + T3 groups are distinguished by the doses of T3 which are givenper 100 g BW under the group name.

differences were observed between hypothyroid rats and3,3'-T2- or 3,5-T2-treated animals, whatever the dose (datanot shown).

Calculation of the protein/DNA ratio (see Table 5) revealedthat chronic T3 administration induced a hypertrophic

effect in the liver and heart and a hyperplastic effect inmuscle and BAT. The evidence for this was as follows. Inliver and heart, T3exerted a strong effect on protein contentand a weaker effect (heart) or no effect (liver) on DNAcontent. The opposite was true for gastrocnemius muscle

J Phy8iol.505.2 535

) by guest on February 25, 2013jp.physoc.orgDownloaded from J Physiol (

M. Moreno, A. Lanni, A. Lombardi and F Goglia

and BAT (i.e. T3 exerted a stronger effect on DNA contentthan on protein content).

DISCUSSIONT3 is generally accepted to be the active form of the variousthyroid hormones and some data seem to indicate that T3may exert both long- and short-term effects on themetabolism of most tissues. However, the question ofwhether T3 has a direct action on mitochondria is stillunresolved. In fact, in spite of several reports that purportto show rapid effects of T3 on mitochondria, the veryoccurrence, and the physiological significance, of theseeffects has long been controversial. As mentioned inIntroduction, this could be due to the great variety ofexperimental conditions used in different laboratories; inparticular, the status of deiodinase enzymes has not alwaysbeen adequately taken into account. This could have led toan effect being attributed to T4 or T3 when it was actuallyattributable to iodothyronines other than T4 and T3.

We have recently shown that isomers of T2 (3,3'-T2 and 3,5-T2) induce a dose-dependent calorigenic effect when injectedinto hypothyroid rats (Lanni et al. 1996). In this presentpaper, we tried to differentiate the effects of T3 and T2. Themost noteworthy result of this study is the finding that T2 shave a similar, but more rapid, effect on RM than T3. Infact, the effect of T2s starts rapidly, reaching a roughly 20%stimulatory effect in 12-15 h, while the effect of T3 startsmore slowly and reaches a roughly 20% stimulatory effectonly after about 50 h.

The need to use a different method of treatment for 3,3'-T2than for 3,5-T2 in order to show up its effect may be relatedto the different behaviour of the metabolic pathwaysinvolved in the degradation (metabolism) of 3,3'-T2 and 3,5-T2 (deiodination, glucuronidation and sulphation). Under ourconditions, deiodination is almost completely blocked (seeTable 2) and, that being so, sulphation and glucuronidationwill be responsible for metabolizing these T2s. In rat liver,the rate of glucuronidation is about 20 times faster for3,3'-T2 than for 3,5-T2 (Visser et al. 1993), and thesulphation rate is immensely faster for 3,3'-T2 than for 3,5-T2 (700 nmol min-' (mg protein)- for 3,3'-T2 and about zerofor 3,5-T2; Sekura, Sato, Cahnmann, Robbins & Jakoby,1981).

The different time course of the effects exerted by T2s andT3on RM may be an indication that they play different rolesat a cellular level. The effect exerted by T2s is too rapid toinvolve protein synthesis, as confirmed by experiments inthe present study in which we used actinomycin D to inhibitprotein synthesis. The simultaneous injection ofactinomycin D with T2s failed to cause any attenuation ofthe stimulation seen with T2s alone. In contrast, whenactinomycin D and T3 were given together, the stimulationof RM by T3 was almost completely abolished (except for asmall effect after 5 days).

While the very rapid effect of T2s (and the lack of anyinhibition of their effect by actinomycin D) could beexplained by their direct interaction with some componentsof the energy releasing apparatus - and the presence ofspecific binding sites for T2s on cell organelles is inaccordance with this notion - the effect of T3 on RM couldbe mediated by a slower action on protein synthesis. Thiseffect of T3 is widely accepted. Indeed, it is well establishedthat the 'long-term' effects of T3 are exerted via actions ontranscription and translation that are triggered by itsbinding to specific nuclear c-erb-A-related receptor proteins(for reviews see Oppenheimer, Schwartz, Mariash, Kinlaw,Wong & Freke, 1987; Lazar, 1993). However, it has still tobe clarified which proteins are 'controlled' by T3.

We wondered whether, in addition to the effect it exerts onthe synthesis of particular mitochondrial components(Scarpulla, Kilar & Scarpulla, 1986; Luciakova & Nelson,1992; Dumnler, Muller & Seitz, 1996), T3 might regulate thecellularity (hypertrophy or hyperplasy) of those organs thatare metabolically very active (such as liver, BAT, muscle andheart). To clarify this point, we chronically treated somegroups of animals with either T3 or T2s. Following suchtreatment, we observed a stimulation of organ growth onlyafter the administration of T3. Moreover, a major part of theincrease in oxidative capacity in a given tissue (total COX)appeared to be secondary to the hypertrophic or hyper-plastic effect exerted by T3. In fact, T3 administrationinduced a hypertrophic effect in the liver and heart, but ahyperplastic effect in muscle and in BAT. As a consequence,and in contrast to the situation seen with T2s, thepercentage increase in COX activity in T3-treated rats wasconsiderably greater when expressed in terms of total tissueactivity than when expressed in terms of specific activity.This analysis of the effects of T3 also supplies an explanationfor the apparent discrepancy between its sustained effect onRM and its weak stimulating effect on specific COX activityat tissue level when given chronically (as seen in a previouswork (Lanni et al. 1996) and as mentioned in Introduction).

In conclusion, we propose that the data presented in thispaper provide an answer to the problem of how iodo-thyronines are involved in the regulation of metabolicactivity by the thyroid. According to our hypothesis, T3exerts an important part of its effect on metabolic rate bymodulating the cellularity of those organs that aremetabolically very active. This effect, inducing anenhancement of the percentage contribution bymetabolically very active tissues to the total body weight(see Table 4), would increase the animal's oxygenconsumption at rest. Owing to the involvement of themachinery of protein synthesis, this effect would take somedays to start and some days to stop. On the other hand, T2swould exert their effect more rapidly than T3 and, aspreviously hypothesized (Lanni et al. 1992, 1993, 1994a,b,1996), this effect would be mediated by a direct interactionbetween diiodothyronines and mitochondria, with 3,5-T2seeming to show a clearer effect. In normal animals, these

J Phy8iol. 505.2536

) by guest on February 25, 2013jp.physoc.orgDownloaded from J Physiol (

actions would probably not be mutually exclusive, insteadthey might co-operate in determining the final metabolicstate of the organs and, consequently, of the whole animal.In other words, they might co-operate in establishing the'normal level' of the animal's metabolic rate.

ACHESON, K. J. & BURGER, A. J. (1980) A study of the relationshipbetween thermogenesis and thyroid hormones. Journal of Clinicaland Endocrinological Metabolism 51, 84-89.

AULIE, A. & GRAV, H. J. (1979). Effect of cold acclimation on theoxidative capacity of skeletal muscles and liver in young bantamchicks. Comparative Biochemistry and Physiology 62A, 335-338.

BARRE, H., BAILLY, L. & ROUANET, J. L. (1987). Increased oxidativecapacity in skeletal muscle from cold-acclimated ducklings: a

comparison with rats. Comparative Biochemistry and Physiology88B, 519-522.

BIANCO, A. C & SILVA, J. E. (1987). Optimal response of key enzymes

and uncoupling protein to cold in BAT depends on local T3generation. American Journal of Physiology 253, E255-263.

BURTON, K. (1956). A study of the conditions and mechanism of thediphenylamine reaction for the colorimetric estimation ofdeoxyribonucleic acid. Biochemical Journal 62, 315-323.

DUMMLER, K., MULLER, S. & SEITZ, H. J. (1996). Regulation ofadenine translocase and glycerol 3-phosphate dehydrogenaseexpression by thyroid hormones in different rat tissues. BiochemicalJournal 317, 913-919.

ESFANDIARI, A., COURTIN, F., LENNON, A. M., GAVARET, J. M. &PIERRE, M. (1992). Induction of type III deiodinase activity inastroglial cells by thyroid hormones. Endocrinology 131,1682-1688.

FRANCAVILLA, S., CORDESCHI, G., PROPERZI, G., Di Cicco, L.,JANNINI, E. A., PALMERO, S., FUGASSA, E., LoRAs, B. &D'ARMIENTO, M. (1991). Effect of thyroid hormone on the pre- andpost-natal development of the rat testis. Journal of Endocrinology129, 35-42.

GOGLIA, F., LANNI, A., HORST, C., MORENO, M. & THOMA, R. (1994).In vitro binding of 3,5-diiodo-L-thyronine to rat liver mitochondria.Journal of Molecular Endocrinology 13, 275-282.

GOLDBERG, I. H. & RABINOWITZ, M. (1962). Actinomycin D inhibitionof deoxyribonucleic acid-dependent synthesis of ribonucleic acid.Science 136, 315-316.

HARTREE, E. F. (1972). Determination of protein: a modification ofthe Lowry method that gives a linear photometric response.

Analytical Biochemistry 48, 422-427.

JUMP, B. D., NARAYAN, P., TOWLE, H. & OPPENHEIMER. J. H. (1984).Rapid effect of triiodothyronine on hepatic gene expression.Hybridization analysis of tissue-specific triiodothyronine regulationofmRNA S14. Journal of Biological Chemistry 259, 2789-2797.

KAPLAN, M. M. & UTIGER, R. D. (1978). lodothyronine metabolism inliver and kidney homogenates from hyperthyroid and hypothyroidrats. Endocrinology 103, 156-163.

KOHRLE, J. (1994). Thyroid hormone deiodination in target tissues, a

regulatory role for the trace element selenium? Experimental andClinical Endocrinology 102, 63-89.

LANNI, A., MORENO, M., CIOFFI, M. & GOGLIA, F. (1992). Effect of3,3'-diiodothyronine and 3,5-diiodothyronine on rat liver oxidativecapacity. Molecular and Cellular Endocrinology 86, 143-148.

537

LANNI, A., MORENO, M., CIOFFI, M. & GOGLIA, F. (1993). Effect of3,3'-diiodothyronine and 3,5-diiodothyronine on rat livermitochondria. Journal of Endocrinology 136, 59-64.

LANNI, M., MORENO, M., HORST, C., LOMBARDI, A. & GOGLIA, F.(1994a). Specific binding sites for 3,3'-diiodothyronine (3,3'-T2) inrat liver mitochondria. FEBS Letters 351, 237-240.

LANNI, A., MORENO, M., LOMBARDI, A. & GOGLIA, F. (1994b). Rapidstimulation in vitro of rat liver cytochrome oxidase activity by 3,5-diiodothyronine and by 3,3'-diiodothyronine. Molecular andCellular Endocrinology 99, 89-94.

LANNI, A., MORENO, M., LOMBARDI, A. & GOGLIA, F. (1996).Calorigenic effect of diiodothyronines in the rat. Journal ofPhysiology 494, 831-837.

LAZAR, M. A. (1993). Thyroid hormone receptors: multiple forms,multiple possibilities. Endocrine Review 14, 184-193.

LEONARD, J. L., KAPLAN, M. M., VISSER, T. J., SILVA, J. E. &LARSEN, P. R. (1981). Cerebral cortex responds rapidly to thyroidhormones. Science 214, 571-573.

LEONARD, J. L., MELLEN, S. A. & LARSEN, P. R. (1983). Thyroxine5'-deiodinase activity in brown adipose tissue. Endocrinology 112,1153-1155.

LUcIAKOVA, K. & NELSON, B. D. (1992). Transcript levels for nuclear-encoded mammalian mitochondrial respiratory-chain componentsare regulated by thyroid hormone in an uncoordinated fashion.European Journal of Biochemistry 207, 247-251.

MAGNUS-LEVY, A. (1895). Uber den respiratorischen Gaswechechselunter dem Einfluss der Thyroidea sowie unter verschiedenenpathologischen Zustanden. Berlin Klinische Wochenschrift 32,650-652.

MORENO, M., LANNI, A., LOMBARDI, A. & GOGLIA, F. (1996). Differentmechanisms underly the effects of triiodothyronine anddiiodothyronines on energy metabolism. 23rd Annual Meeting ofthe European Thyroid Association. Amsterdam, The Netherlands.Journal of Endocrinological Investigation 19, suppl. 6, 47.

OPPENHEIMER, J. H., SCHWARTZ, H. L., LANE, J. T. & THOMPSON,M. P. (1991). Functional relationship of thyroid hormone-inducedlipogenesis, lipolysis, and thermogenesis in the rat. Journal ofClinical Investigation 87, 125-132.

OPPENHEIMER, J. H., SCHWARTZ, H. L., MARIASH, C. N., KINLAW,W. B., WONG, N. C. W. & FREKE, H. C. (1987). Advances in our

understanding of thyroid hormone action at the cellular level.Endocrinological Reviews 8, 288-308.

O'REILLY, I. & MURPHY, M. P. (1992). Studies on the rapid stimulationof mitochondrial respiration by thyroid hormones. ActaEndocrinologica 127, 542-546.

SCARPULLA, R. C., KILAR, M. C. & SCARPULLA, K. M. (1986).Coordinate induction of multiple cytochrome c mRNAs in response

to thyroid hormones. Journal of Biological Chemistry 261,4660-4662.

SCHOENMAKERS, C. H. H., PIGMANS, I. G. & VISSER, T. J. (1995).Investigation of type I and type III iodothyronine deiodinases inrat tissues using N-bromoacetyl-iodothyronine affinity labels.Molecular and Cellular Endocrinology 107, 173-180.

SEKURA, R. D., SATo, K., CAHNMANN, H. J., ROBBINS, J. & JAKOBY,W. B. (1981). Sulfate transfer to thyroid hormones and their analogsby hepatic aryl sulfotransferases. Endocrinology 108, 454-456.

SESTOFT, L. (1980). Metabolic aspects of calorigenic effect of thyroidhormone in mammals. Clinical Endocrinology (Oxford) 13, 489-512.

SOBOLL, S. (1993). Thyroid hormone action on mitochondrial energy

transfer. Biochimica et Biophysica Acta 1144, 1-16.

J Physiol. 505.2 Iodothyronines and metabolic rate

) by guest on February 25, 2013jp.physoc.orgDownloaded from J Physiol (

M. Moreno, A. Lanni, A. Lombardi and F Goglia

TATA, J. R. (1963). Inhibition of the biological action of thyroidhormones by actinomycin D and puromycin. Nature 197,1167-1168.

TATA, J. R., ERNSTER, L. & LINDBERG, 0. (1962). Control of basalmetabolic rate by thyroid hormones and cellular function. Nature193, 1058-1060.

TATA, J. R., ERNSTER, L., LINDBERG, O., ARRHENIUS, E., PEDERSEN,S. & HEDMAN, R. (1963). The action of thyroid hormone at the celllevel. Biochemical Journal 86, 408-428.

VISSER, T. J., DOCTER, R. & HENNEMANN, G, (1977). Radioimmuno-assay of reverse triiodothyronine. Journal of Endocrinology 73,395-396.

VISSER, T. J., KAPTEIN, E., TERPSTRA, 0. T. & KRENNING, E. P.(1988). Deiodination of thyroid hormone by human liver. Journal ofClinical and Endocrinological Metabolism 67, 17-24.

VISSER, T. J., KRIEGER-QUIST, L. M., DOCTER, R. & HENNEMANN, G.(1978). Radioimmunoassay of 3,3'-diiodothyronine in unextractedserum: the effect of endogenous triiodothyronine. Journal ofEndocrinology 79, 357-361.

VISSER, T. J., MORENO, M., HuISWOULD, G. A., GOGLIA, F., THOMA,R., HORST, C., EBNER, T. & BURCHELL, B. (1993). Structure-activityrelationship of iodothyronine glucuronidation by human and ratUDP-glucuronyltransferases (UGTs). 67th Meeting AmericanThyroid Association, Tampa, FL, USA. Thyroid T-5.2.

AcknowledgementsThis work was funded by the European Commission (contract no.ERBCHRX-CT940490) and by Ministero dell'Universita e dellaRicerca Scientifica e Tecnologica (40% and 60 %).

Received 18 June 1997; accepted 8 August 1997.

538 J Phy8iol.505.2

) by guest on February 25, 2013jp.physoc.orgDownloaded from J Physiol (